OBJECTIVE: To evaluate the sensitivity and specificity of Doppler ultrasound (DUS) in diagnosing arthritis in the wrist and hands, and, if possible, to define a cutoff level for our ultrasound measures for inflammation, resistive index (RI), and color fraction. METHODS: Using DUS, 88 patients with active RA were selected for study and 27 healthy controls. A total of 419 joints were examined. The synovial vascularization was determined by color Doppler and spectral Doppler estimating the color fraction (the percentage of color pixels inside the synovium was the region of interest) and RI in wrist, metacarpophalangeal (MCP), and proximal interphalangeal (PIP) joints. Receiver-operator characteristic (ROC) curves were made for both US measures. Cutoff levels were selected from the ROC curves as the values with the optimum sensitivity and specificity. RESULTS: Analyses were carried out for small joints (MCP and PIP), wrists, and for all joints (pooled). Pooled joint analysis showed the area under the curve for both RI and color fraction was 0.84. The cutoff level for the color fraction was 0.01 and for RI 0.83. With these cutoff levels, the sensitivity and specificity for the color fraction were 0.92 and 0.73, respectively. For RI a sensitivity of 0.72 and specificity of 0.70 were found. Analysis of small joints and wrist gave very similar results. CONCLUSION: DUS may detect vascularization of the inflamed synovium with a high sensitivity and a moderate specificity with selected cutoff levels.
OBJECTIVE: To evaluate the sensitivity and specificity of Doppler ultrasound (DUS) in diagnosing arthritis in the wrist and hands, and, if possible, to define a cutoff level for our ultrasound measures for inflammation, resistive index (RI), and color fraction. METHODS: Using DUS, 88 patients with active RA were selected for study and 27 healthy controls. A total of 419 joints were examined. The synovial vascularization was determined by color Doppler and spectral Doppler estimating the color fraction (the percentage of color pixels inside the synovium was the region of interest) and RI in wrist, metacarpophalangeal (MCP), and proximal interphalangeal (PIP) joints. Receiver-operator characteristic (ROC) curves were made for both US measures. Cutoff levels were selected from the ROC curves as the values with the optimum sensitivity and specificity. RESULTS: Analyses were carried out for small joints (MCP and PIP), wrists, and for all joints (pooled). Pooled joint analysis showed the area under the curve for both RI and color fraction was 0.84. The cutoff level for the color fraction was 0.01 and for RI 0.83. With these cutoff levels, the sensitivity and specificity for the color fraction were 0.92 and 0.73, respectively. For RI a sensitivity of 0.72 and specificity of 0.70 were found. Analysis of small joints and wrist gave very similar results. CONCLUSION: DUS may detect vascularization of the inflamed synovium with a high sensitivity and a moderate specificity with selected cutoff levels.
Authors: Carlos Acebes; Neil McKay; Anna Ciechomska; Nicola Alcorn; John P Harvie; Barbara Robson; Nico Groenendijk; Moira McDonald; Alison Wilson; Jesus Garrido Journal: Rheumatol Int Date: 2016-11-28 Impact factor: 2.631